Endobronchial valves placement for persistent air leaks in critically ill patients: chest tube removal and competing risk of death.

Background: Persistent air leaks due to alveolar-pleural or peripheral bronchiolar-pleural fistulas remain challenging in critically ill patients requiring respiratory support. We assessed the effectiveness and safety of endobronchial valve placement in this setting, accounting for the competing risk of death. Methods: We conducted a single-center, retrospective cohort study of consecutive critically ill adults requiring invasive mechanical ventilation or high-flow nasal oxygen for acute respiratory failure who underwent endobronchial valve placement for alveolar-pleural or peripheral bronchiolar-pleural fistulas. The primary outcome was time to chest tube removal, analyzed using a competing risk analysis in which in-hospital death was treated as the competing event. Secondary outcomes included in-hospital mortality, procedural safety, and exploratory analyses of the response to early post-procedural air leaks. Results: A total of 38 patients were included. The cumulative incidence of chest tube removal reached 60.5, 71.1, and 94.7% by days 10, 14, and 20, respectively, after endobronchial valve placement; the corresponding cumulative incidence of in-hospital death was 2.6% by day 14. The median time to chest tube removal was 7.5 days. In Fine-Gray models, adjusted for clinically relevant covariates, older age was found to be associated with a lower incidence of chest tube removal (subdistribution hazard ratio [sHR], 0.96 per year; 95% CI, 0.93-0.98; Conclusion: In critically ill patients with persistent air leaks, endobronchial valve placement was associated with rapid chest tube removal and a low early competing risk of death. These findings support further prospective evaluation of this approach in high-risk patients.